Now that a short-term path for incremental improvements to the next EU-wide stress test exercise has started, it is time to reflect on more fundamental changes that could be made to the stress testing framework within the EU. Stress tests have become an important part of the supervisory toolkit. Increasing their effectiveness and efficiency is therefore not only desirable, but also essential to better accomplish our task in this field.
This paper argues that significant benefits can be achieved by decoupling micro-prudential stress tests from the oversight/macro-surveillance ones. Decoupling the two is not just legally possible, but also consistent with the different tasks assigned to the relevant authorities in the field of stress testing: on one side, Art. 100 CRDIV for supervisory stress tests on institutions ('micro-prudential exercises') and, on the other side, the EBA Regulation, which requires the EBA to initiate and coordinate EU-wide stress tests with the aim of assessing the resilience of financial institutions and contributing to the overall assessment of systemic risk in the EU financial system ('macro-prudential exercises'). Having two separate exercises for two different goals (rather than one exercise for two different goals) would be more effective and cost-efficient, for both authorities and banks, to the extent the design of the exercises is more closely related to their policy goals.
Micro-prudential stress tests for supervisory purposes (i.e. for the SREP process) should be: a) more robust and targeted, with a focus on specific risk areas and shocks identified according tothe supervisory priorities; b) better balanced between static and dynamic assumptions and more inclusive of the ICAAP process; c) mainly bottom-up, under the control of the competent authority; and d) not necessarily for disclosure.
Macro-prudential stress tests for oversight/macro surveillance goals should be aimed at assessing the stability and soundness of the EU financial sector and its capacity to provide financing to the real economy. They should mainly be top-down and fairly flexible in terms, for example, of the number of scenarios and the inclusion of emerging risks. The results obtained in this way would ensure a consistent assessment of system-wide resilience by providing an aggregate envelope for capital depletion, with some of the aggregate figures made public although possibly without reference to individual banks.
